Cancer Treatments with Big Benefits for Small Populations

June 13, 2010 - 12:34 pm email to a friend

Print This Post view / write comments

(6 votes, average: 5.00 out of 5)

Loading ...

Dr West

Earlier this week, I wrote about the very promising early work with the novel, still investigational agent crizotinib (PF-02341066) in treating the 4-5% of NSCLC patients with a tumor positive for an ALK rearrangement. For this minority of patients, the benefit was of a kind we don’t see enough in cancer and certainly not in lung cancer, with about 2/3 of patients showing a significant response in terms of tumor shrinkage, and 87% achieving at least stable disease; nearly 3 out of 4 remained on it without progression 6 months later, and many are on it now for much longer, with responses still ongoing.

These data, presented in the Plenary Session of our ASCO Annual Conference, were really not controversial, except for the question of whether crizotinib should be approved by the FDA (and potentially by other similar regulators in countries around the world) based on these overwhelmingly positive results, in a limited population. The larger studies are ongoing that I think are exceptionally likely to lead to its official FDA approval, but in the meantime, there is no question that many patients will miss the opportunity to benefit from it while it remains available in only very selected settings around the world. There will also be less awareness of the test for ALK rearrangements and the benefits of crizotinib in the appropriate target patients while it is not commercially available and marketed.

The biggest problem of all, though, is that having a very effective treatment for a limited segment of the population creates a whole new set of problems that we’ll see more and more. In many ways, we are fortunate to have found the right target for the right targeted therapy here. If crizotinib had been tested in a broad NSCLC population, as is the ritual default paradigm for our treatments, it would likely be just one more drug with a 3-4% response rate that would be discarded rather than progressing through additional clinical testing. How many other crizotinibs are there that just haven’t had their appropriate targets identified and remain undeveloped by a drug company because they appear to be minimally active in an overly broad population? We should also remember that many, many agents have now been tested in too broad a population and had safety problems. Many of the anti-angiogenic agents have been shown to have increased mortality compared with placebo in patients with squamous NSCLC; in fact, it’s easy to imagine that Avastin (bevacizumab) would be tainted as the drug that regularly kills people by hemoptysis if investigators hadn’t identified the problem early in its development in lung cancer (as reviewed in this post) and excluded patients with squamous NSCLC in the larger phase III studies that led to its ultimate approval.

But we should also be concerned that pharmaceutical and biotech companies will do the calculations of offering treatment to a smaller population and find that it isn’t worth their costs and efforts to pursue therapies for a small subset of patients. Here is one of the first slides from Dr. Pennell’s excellent podcast on molecular therapies for NSCLC, showing that until 10 years ago, NSCLC was essentially viewed as one entity with a market of 60-70,000 patients with advanced disease (and smaller numbers with other stages as well) in the US each year:

(click on image to enlarge)

Treatments were given in this way, and in the second half of the 1990s, carbo/taxol (paclitaxel) was the overwhelmingly used doublet in first line therapy, and the company making it (Bristol-Myers Squibb) was like the British Empire of oncology. There was little question of the value of having a large stake in that market.

Dr. Pennell, toward the end of his podcast, contrasted that old view with a more current perspective and future projection of the field, which is far more segmented. In fact, his 2020 projection includes no large populations, but rather molecularly defined small groups that presumably can achieve greater benefits than today by receiving a more personalized treatment approach.

This new oncologic world order of smaller populations achieving greater benefits, however, has no blockbuster treatments. We can only hope that there are enough incentives for pharma/biotech companies to develop agents that will be ideal treatments for populations of mere thousands or perhaps even only hundreds of patients instead of tends of thousands. Perhaps companies will seek approval for their drugs based on a limited subset and hope that oncologists will also offer it to broader populations once it’s commercially available.

It would certainly help if the FDA changes its requirements for getting drugs approved in smaller populations. Agents that may be very helpful in populations that amount to only hundreds or a few thousand patients per year can’t feasibly be tested in proper randomized prospective phase III studies (see here for review of basics of drug development and types of trials). A shorter and less arduous drug approval process will greatly help reduce the costs and other disincentives for a company to bring a novel agent to market for a relatively small target population/market.

Reaching the right patients when they are uncommon to rare and geographically dispersed creates other challenges. All of these molecularly defined therapies are predicated on having tumor tissue available for testing, far more than we’ve routinely sought just to make a diagnosis. And the onus may be much more on patients to stay on top of opportunities and travel to pursue trials that are seeking their particular cancer subtype.

More than ever before, the pace of clinical research and drug approvals may depend on having a knowledgeable, empowered patient population who can help identify the best trial opportunities for themselves.

I welcome your thoughts.

Bookmark to: